New recipe NZZ Webpaper

[skoll1975]

Hi Kovid,

please find attached the revised recipe for NZZ Webpaper (payed version).

Obviously there was a recent change in src/calibre/ebook/oeb/parse_utils that added additional syntax checks for HTML. They broke my recipe because it used incomplete HTML files that were not accepted by parse_utils after the change (shame on me ...).

So the new version delivers more correct HTML than the old one.

In detail: the function parse_html is now looking for a string "<html" - if it doesnt find one, the call of check_for_html5 fails in line 276 because of undefined value in variable pre. Maybe you want to add some additional checks to handle this condition.

Regards,

Bernd

[kovidgoyal]

updated.

[gr650]

Does anyone have an idea why the recipe doesn't download the article images? I get a nicely navigable epub, but without the images (which are there in the webpaper version). I'm running the ebook-convert commandline conversion.

Just automated the epub creation to run in the morning, putting the epub file straight into my dropbox: http://chruss.wordpress.com/2014/05/...-ebook-reader/

[skoll1975]

Hi gr650,

I deliberately removed the images from the webpaper, because I wanted a small epub result (think about the 100 mb the official download packages have) and greyscale images are not displayed nicely on my kindle anyway.

So, it's not a bug, it's a feature.

Regards,

Bernd